def transform(
self,
object_to_transform: Union[io.BufferedIOBase, io.RawIOBase,
AnyPDFType],
parent_object: Any,
context: Optional[ReadTransformerContext] = None,
event_listeners: typing.List[EventListener] = [],
) -> Any:
"""
This function writes a \FontDescriptor Dictionary to a byte stream
"""
assert isinstance(object_to_transform, Dictionary)
# convert like regular dictionary
if isinstance(parent_object, Font):
for t in self.get_root_transformer().children:
if isinstance(t, ReadDictionaryTransformer):
return t.transform(object_to_transform, parent_object,
context, event_listeners)
# build intermittent Font object
tmp = Font().set_parent(parent_object) # type: ignore [attr-defined]
# add listener(s)
for l in event_listeners:
tmp.add_event_listener(l)
tmp[Name("FontDescriptor")] = self.get_root_transformer().transform(
object_to_transform, tmp, context, [])
# return
return tmp
def transform(
self,
object_to_transform: Union[io.BufferedIOBase, io.RawIOBase,
AnyPDFType],
parent_object: Any,
context: Optional[TransformerContext] = None,
event_listeners: typing.List[EventListener] = [],
) -> Any:
# convert like regular dictionary
if isinstance(parent_object, Font):
for t in self.get_root_transformer().handlers:
if isinstance(t, DefaultDictionaryTransformer):
return t.transform(object_to_transform, parent_object,
context, event_listeners)
# build intermittent Font object
tmp = Font().set_parent(parent_object) # type: ignore [attr-defined]
# add listener(s)
for l in event_listeners:
tmp.add_event_listener(l)
tmp["FontDescriptor"] = self.get_root_transformer().transform(
object_to_transform, tmp, context, [])
# return
return tmp
def __init__(
self,
text_bytes: bytes,
font: Font,
font_size: Decimal,
character_spacing: Decimal = Decimal(0),
word_spacing: Decimal = Decimal(0),
horizontal_scaling: Decimal = Decimal(100),
):
assert isinstance(font, Font)
self._glyphs: typing.List[Glyph] = []
i: int = 0
while i < len(text_bytes):
# sometimes, 2 bytes make up 1 unicode char
unicode_chars: typing.Optional[str] = None
if i + 1 < len(text_bytes):
multi_byte_char_code: int = text_bytes[i] * 256 + text_bytes[
i + 1]
unicode_chars = font.character_identifier_to_unicode(
multi_byte_char_code)
if unicode_chars is not None:
self._glyphs.append(
Glyph(
multi_byte_char_code,
unicode_chars,
font.get_width(multi_byte_char_code)
or pDecimal(0),
))
i += 2
continue
# usually it's 1 byte though
if i < len(text_bytes):
unicode_chars = font.character_identifier_to_unicode(
text_bytes[i])
if unicode_chars is not None:
self._glyphs.append(
Glyph(
text_bytes[i],
unicode_chars,
font.get_width(text_bytes[i]) or Decimal(0),
))
i += 1
continue
# no mapping found
if i < len(text_bytes):
self._glyphs.append(Glyph(text_bytes[i], "�", Decimal(250)))
i += 1
self._font = font
self._font_size = font_size
self._character_spacing = character_spacing
self._word_spacing = word_spacing
self._horizontal_scaling = horizontal_scaling
def _find_best_matching_predefined_cmap(
cmap_name: str) -> typing.Dict[int, str]:
cmap_dir: Path = Path(__file__).parent / "cmaps"
assert cmap_dir.exists()
predefined_cmaps: typing.List[str] = [
x.name for x in cmap_dir.iterdir()
]
# pseudo match
if cmap_name not in predefined_cmaps:
if cmap_name == "Adobe-Identity-UCS2":
logger.info(
"Encoding Adobe-Identity-UCS2 was specified, using Adobe-Identity-H in stead"
)
cmap_name = "Adobe-Identity-H"
if cmap_name == "Adobe-Japan1-UCS2":
logger.info(
"Encoding Adobe-Identity-UCS2 was specified, using Adobe-Japan1-0 in stead"
)
cmap_name = "Adobe-Japan1-0"
if cmap_name not in predefined_cmaps:
logger.info(
"Encoding %s was specified, defaulting to Adobe-Identity-H in stead"
% cmap_name)
cmap_name = "Adobe-Identity-H"
cmap_bytes: typing.Optional[bytes] = None
with open(cmap_dir / cmap_name, "rb") as cmap_file_handle:
cmap_bytes = cmap_file_handle.read()
assert cmap_bytes is not None
return Font._read_cmap(cmap_bytes)
def _read_file(input: io.IOBase) -> Optional[Font]:
lines = [x for x in input.readlines() if not x.startswith("Comment")]
lines = [x[:-1] if x.endswith("\n") else x for x in lines]
# check first/last line
if not lines[0].startswith("StartFontMetrics") or not lines[-1].startswith(
"EndFontMetrics"
):
return None
out_font = Font()
# FontDescriptor
out_font_descriptor = FontDescriptor().set_parent(out_font)
out_font_descriptor["FontName"] = AdobeFontMetrics._find_and_parse_as_string(
lines, "FontName"
)
out_font_descriptor["FontFamily"] = AdobeFontMetrics._find_and_parse_as_string(
lines, "FamilyName"
)
# FontStretch
# FontWeight
# Flags
# FontBBox
# ItalicAngle
out_font_descriptor["Ascent"] = AdobeFontMetrics._find_and_parse_as_float(
lines, "Ascender"
)
out_font_descriptor["Descent"] = AdobeFontMetrics._find_and_parse_as_float(
lines, "Descender"
)
# Leading
out_font_descriptor["CapHeight"] = AdobeFontMetrics._find_and_parse_as_float(
lines, "CapHeight"
)
out_font_descriptor["XHeight"] = AdobeFontMetrics._find_and_parse_as_float(
lines, "XHeight"
)
# StemV
out_font_descriptor["StemV"] = AdobeFontMetrics._find_and_parse_as_float(
lines, "StemV"
)
# StemH
out_font_descriptor["StemH"] = AdobeFontMetrics._find_and_parse_as_float(
lines, "StemH"
)
# AvgWidth
out_font_descriptor["AvgWidth"] = AdobeFontMetrics._find_and_parse_as_float(
lines, "AvgWidth"
)
# MaxWidth
out_font_descriptor["MaxWidth"] = AdobeFontMetrics._find_and_parse_as_float(
lines, "MaxWidth"
)
# MissingWidth
out_font_descriptor["MissingWidth"] = AdobeFontMetrics._find_and_parse_as_float(
lines, "MissingWidth"
)
# FontFile
# FontFile2
# FontFile3
out_font_descriptor["CharSet"] = AdobeFontMetrics._find_and_parse_as_integer(
lines, "Characters"
)
# Font
out_font["Type"] = "Font"
out_font["Subtype"] = "Type1"
out_font["Name"] = out_font_descriptor["FontName"]
out_font["BaseFont"] = out_font_descriptor["FontName"]
widths = List().set_parent(out_font)
avg_char_width = 0
avg_char_width_norm = 0
first_char = None
last_char = None
char_metrics_lines = lines[
lines.index(
[x for x in lines if x.startswith("StartCharMetrics")][0]
) : lines.index("EndCharMetrics")
+ 1
]
char_metrics_lines = char_metrics_lines[1:-1]
for cml in char_metrics_lines:
tmp = {
y.split(" ")[0]: y.split(" ")[1]
for y in [x.strip() for x in cml.split(";")]
if " " in y
}
# determine char
ch = -1
if "C" in tmp:
ch = int(tmp["C"])
if "CH" in tmp:
ch = int(tmp["CH"][1:-1], 16)
if (first_char is None or ch < first_char) and ch != -1:
first_char = ch
if (last_char is None or ch > last_char) and ch != -1:
last_char = ch
w = float(tmp["WX"])
if ch != -1 and w != 0:
avg_char_width += w
avg_char_width_norm += 1
widths.append(Decimal(w))
out_font["FirstChar"] = Decimal(first_char)
out_font["LastChar"] = Decimal(last_char)
out_font["Widths"] = widths
if out_font_descriptor["AvgWidth"] is None:
out_font_descriptor["AvgWidth"] = round(
Decimal(avg_char_width / avg_char_width_norm), 2
)
if out_font_descriptor["MaxWidth"] is None:
out_font_descriptor["MaxWidth"] = max(widths)
out_font["FontDescriptor"] = out_font_descriptor
# return
return out_font
def _read_file(input: typing.TextIO) -> Optional[Font]:
lines: typing.List[str] = [x for x in input.readlines()]
lines = [x for x in lines if not x.startswith("Comment")]
lines = [x[:-1] if x.endswith("\n") else x for x in lines]
# check first/last line
if not lines[0].startswith("StartFontMetrics") or not lines[-1].startswith(
"EndFontMetrics"
):
return None
out_font = Font()
# FontDescriptor
out_font_descriptor = FontDescriptor().set_parent(out_font) # type: ignore [attr-defined]
font_name = AdobeFontMetrics._find_and_parse_as_string(lines, "FontName")
if font_name:
out_font_descriptor[Name("FontName")] = Name(font_name)
font_family = AdobeFontMetrics._find_and_parse_as_string(lines, "FamilyName")
if font_family:
out_font_descriptor[Name("FontFamily")] = String(font_family)
# FontStretch
# FontWeight
# Flags
# FontBBox
fontbbox_str = AdobeFontMetrics._find_and_parse_as_string(lines, "FontBBox")
if fontbbox_str:
fontbbox = [Decimal(x) for x in fontbbox_str.split(" ")]
out_font_descriptor[Name("FontBBox")] = List().set_can_be_referenced(False) # type: ignore [attr-defined]
for x in fontbbox:
out_font_descriptor[Name("FontBBox")].append(x)
# ItalicAngle
italic_angle = AdobeFontMetrics._find_and_parse_as_float(lines, "ItalicAngle")
if italic_angle:
out_font_descriptor[Name("ItalicAngle")] = Decimal(italic_angle)
else:
out_font_descriptor[Name("ItalicAngle")] = Decimal(0)
# Ascent
ascent = AdobeFontMetrics._find_and_parse_as_float(lines, "Ascender")
if ascent:
out_font_descriptor[Name("Ascent")] = Decimal(ascent)
else:
out_font_descriptor[Name("Ascent")] = Decimal(0)
# Descent
descent = AdobeFontMetrics._find_and_parse_as_float(lines, "Descender")
if descent:
out_font_descriptor[Name("Descent")] = Decimal(descent)
else:
out_font_descriptor[Name("Descent")] = Decimal(0)
# Flags
out_font_descriptor[Name("Flags")] = Decimal(131104)
# Leading
# CapHeight
capheight = AdobeFontMetrics._find_and_parse_as_float(lines, "CapHeight")
if capheight:
out_font_descriptor[Name("CapHeight")] = Decimal(capheight)
else:
out_font_descriptor[Name("CapHeight")] = Decimal(0)
# XHeight
xheight = AdobeFontMetrics._find_and_parse_as_float(lines, "XHeight")
if xheight:
out_font_descriptor[Name("XHeight")] = Decimal(xheight)
# StemV
stemv = AdobeFontMetrics._find_and_parse_as_float(lines, "StemV")
if stemv:
assert stemv is not None
out_font_descriptor[Name("StemV")] = Decimal(stemv)
else:
out_font_descriptor[Name("StemV")] = Decimal(0)
# StemH
stemh = AdobeFontMetrics._find_and_parse_as_float(lines, "StemH")
if stemh:
assert stemh is not None
out_font_descriptor[Name("StemH")] = Decimal(stemh)
else:
out_font_descriptor[Name("StemH")] = Decimal(0)
# AvgWidth
avgwidth = AdobeFontMetrics._find_and_parse_as_float(lines, "AvgWidth")
if avgwidth:
assert avgwidth is not None
out_font_descriptor[Name("AvgWidth")] = Decimal(avgwidth)
# MaxWidth
maxwidth = AdobeFontMetrics._find_and_parse_as_float(lines, "MaxWidth")
if maxwidth:
assert maxwidth is not None
out_font_descriptor[Name("MaxWidth")] = Decimal(maxwidth)
# MissingWidth
missingwidth = AdobeFontMetrics._find_and_parse_as_float(lines, "MissingWidth")
if missingwidth:
assert missingwidth is not None
out_font_descriptor[Name("MissingWidth")] = Decimal(missingwidth)
# CharSet
charset = AdobeFontMetrics._find_and_parse_as_float(lines, "CharSet")
if charset:
assert charset is not None
out_font_descriptor[Name("CharSet")] = Decimal(charset)
# Font
out_font[Name("Type")] = Name("Font")
out_font[Name("Subtype")] = Name("Type1")
out_font[Name("Name")] = out_font_descriptor["FontName"]
out_font[Name("BaseFont")] = out_font_descriptor["FontName"]
widths = List().set_parent(out_font) # type: ignore [attr-defined]
avg_char_width: float = 0
avg_char_width_norm: float = 0
first_char = None
last_char = None
char_metrics_lines = lines[
lines.index(
[x for x in lines if x.startswith("StartCharMetrics")][0]
) : lines.index("EndCharMetrics")
+ 1
]
char_metrics_lines = char_metrics_lines[1:-1]
for cml in char_metrics_lines:
tmp = {
y.split(" ")[0]: y.split(" ")[1]
for y in [x.strip() for x in cml.split(";")]
if " " in y
}
# determine char
ch = -1
if "C" in tmp:
ch = int(tmp["C"])
if "CH" in tmp:
ch = int(tmp["CH"][1:-1], 16)
if (first_char is None or ch < first_char) and ch != -1:
first_char = ch
if (last_char is None or ch > last_char) and ch != -1:
last_char = ch
w = float(tmp["WX"])
if ch != -1 and w != 0:
avg_char_width += w
avg_char_width_norm += 1
widths.append(Decimal(w))
assert first_char is not None
assert last_char is not None
out_font[Name("FirstChar")] = Decimal(first_char)
out_font[Name("LastChar")] = Decimal(last_char)
out_font[Name("Widths")] = widths
if avgwidth is None:
out_font_descriptor[Name("AvgWidth")] = Decimal(
round(Decimal(avg_char_width / avg_char_width_norm), 2)
)
if maxwidth is None:
out_font_descriptor[Name("MaxWidth")] = Decimal(max(widths))
out_font[Name("FontDescriptor")] = out_font_descriptor
# return
return out_font